BMC methods最新文献

Background: Accurate assessment of left ventricular function is essential for diagnosing and managing cardiovascular disease. Gated myocardial perfusion SPECT (MPS) enables simultaneous evaluation of perfusion and function, but reliable contour extraction is challenged by image noise, resolution limits, and anatomical variability. Multi-center validation is further restricted by data privacy concerns, underscoring the need for robust and privacy-preserving contouring methods.

Methods: In this study, we propose a novel approach, FedDA-TSformer, which integrates Federated Domain Adaptation with the TimeSformer model for the task of left ventricle segmentation using MPS images. The proposed model captures spatial and temporal features through a Divide-Space-Time-Attention mechanism, which ensures spatial-temporal consistency in predictions across multi-center datasets. To facilitate domain adaptation, we employ a local maximum mean discrepancy (LMMD) loss to align model outputs across data from three different institutions. This strategy effectively combines federated learning and domain adaptation to enhance model generalization while ensuring data security.

Results: We evaluated FedDA-TSformer on a dataset comprising 150 subjects collected from three hospitals, with each cardiac cycle divided into eight gates. The model achieved Dice Similarity Coefficients (DSC) of 0.842 and 0.907 for left ventricular endocardium and epicardium segmentation, respectively.

Discussion: FedDA-TSformer provides a robust, privacy-preserving solution for multi-center left ventricular segmentation, outperforming traditional FedAvg in handling domain shifts. By leveraging the TimeSformer architecture and domain adaptation mechanisms, the framework ensures spatial-temporal consistency and data security across heterogeneous clinical sites. Despite current limitations regarding communication overhead and its focus on a small SPECT-only dataset, this study establishes a scalable foundation for collaborative cardiac diagnosis. Future work will prioritize model compression, asynchronous updates, and cross-modality generalization to CT and MRI to enhance its practicality in resource-constrained environments.

Background: To address the limitations of large-scale high quality microscopy image acquisition, PSSR (Point-Scanning Super-Resolution) was introduced to enhance easily acquired low quality microscopy data to a higher quality using deep learning-based methods. However, while PSSR was released as open-source, it was difficult for users to implement into their workflows due to an outdated codebase, limiting its usage by prospective users. Additionally, while the data enhancements provided by PSSR were significant, there was still potential for further improvement.

Methods: To overcome this, we introduce PSSR2, a redesigned implementation of PSSR workflows and methods built to put state-of-the-art technology into the hands of the general microscopy and biology research community. PSSR2 enables user-friendly implementation of super-resolution workflows for simultaneous super-resolution and denoising of undersampled microscopy data, especially through its integrated Command Line Interface and Napari plugin. PSSR2 improves and expands upon previously established PSSR algorithms, mainly through improvements in the semi-synthetic data generation ("crappification") and training processes.

Results: In benchmarking PSSR2 on a test dataset of paired high and low resolution electron microscopy images, PSSR2 super-resolves high-resolution images from low-resolution images to a significantly higher accuracy than PSSR. The super-resolved images are also more visually representative of real-world high-resolution images.

Discussion: The improvements in PSSR2, in providing higher quality images, should improve the performance of downstream analyses. We note that for accurate super-resolution, PSSR2 models should only be applied to super-resolve data sufficiently similar to training data and should be validated against real-world ground truth data.

Background: The structural dynamics of proteins and nucleic acids are critical for their function in many biological processes but investigating these dynamics is often challenging with traditional techniques. Time-correlated single photon counting (TCSPC) coupled with confocal microscopy is a versatile biophysical tool that enables real-time monitoring of biomolecular dynamics in a variety of systems, across many timescales. Quantitative single-molecule time-resolved fluorescence methods are uniquely positioned to investigate transient interactions and structural changes, yet application in complex biological systems remains limited by technical and analytical challenges. Combining fluorescence lifetime imaging microscopy (FLIM) with pulsed interleaved excitation Förster resonance energy transfer (PIE-FRET) offers a robust approach to overcome these barriers, enabling accurate distance measurements and dynamic studies across diverse sample types.

Methods: We describe practical workflows for implementing FLIM/PIE-FRET for quantitative measurements of nanoscale distances and dynamic processes in various biomolecular systems on a commercial microscope. Benchmark DNA constructs, RNA/DNA hybrids, liposome-encapsulated enzymes, and live Saccharomyces cerevisiae strains were prepared and imaged. Correction factors for FRET efficiency recovery were determined from diffusion-based experiments, and results were validated by direct comparison of intensity- and lifetime-based analyses.

Results: FRET efficiencies from both intensity- and lifetime-based analyses were consistent across systems. DNA standards reproduced expected values, RNA/DNA hybrids reported on substrate dynamics, liposome encapsulation enabled single-enzyme conformational probing, and live-cell imaging revealed transient protein-protein interactions during ribosome biogenesis.

Discussion: This work establishes guidelines for implementing FLIM/PIE-FRET as an accessible method to interrogate nanoscale distances, conformational dynamics, and protein-protein interactions both in vitro and in live cells. The strategies outlined here facilitate broader adoption of quantitative single-molecule time-resolved fluorescence in structural and cell biology.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00048-1.

Background: Macrophages are a promising target for therapeutics in various applications such as regenerative medicine and immunotherapy for cancer. Due to their plastic nature, macrophages can switch from a non-activated state to activated with the smallest environmental change. For macrophages to be effective in their respective applications, screening for phenotypic changes is necessary to elucidate the cell response to different delivery vehicles, vaccines, small molecules, and other stimuli.

Methods: We created a sensitive and dynamic high-throughput screening method for macrophages based on the activation of NF-κB. For this reporter, we placed an mRFP1 fluorescence gene under the control of an inflammatory promoter, which recruits NF-κB response elements to promote expression during the inflammatory response in macrophages. We characterized the inflammatory reporter based on key markers of an inflammatory response in macrophages including TNF-α cytokine release and immunostaining for inflammatory and non-inflammatory cell surface markers. We compared gene delivery and inflammation of several clinically relevant viral vehicles and commercially available non-viral vehicles. Statistical analysis between groups was performed with a one-way ANOVA with post-hoc Tukey's test.

Results: The reporter macrophages demonstrated a dynamic range after LPS stimulation with an EC50 of 0.61 ng/mL that was highly predictive of TNF-α release. Flow cytometry revealed heterogeneity between groups but confirmed population level shifts in pro-inflammatory markers. Finally, we demonstrated utility of the reporter by showing divergent effects with various leading gene delivery vehicles.

Discussion: This screening technique developed here provides a dynamic, high-throughput screening technique for determining inflammatory response by mouse macrophages to specific stimuli. The method presented here provides insight into the inflammatory response in mouse macrophages to different viral and non-viral gene delivery methods and provides a tool for high-throughput screening of novel vehicles.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00030-x.

Background: Collaborative efforts, such as the Human Cell Atlas, are rapidly accumulating large amounts of single-cell data. To ensure that single-cell atlases are representative of human genetic diversity, we need to determine the ancestry of the donors from whom single-cell data are generated. Self-reporting of race and ethnicity, although important, can be biased and is not always available for the datasets already collected.

Methods: Here, we introduce scAI-SNP, a tool to infer ancestry directly from single-cell genomics data. To train scAI-SNP, we identified 4.5 million ancestry-informative single-nucleotide polymorphisms (SNPs) in the 1000 Genomes Project dataset across 3201 individuals from 26 population groups. For a query single-cell dataset, scAI-SNP uses these ancestry-informative SNPs to compute the contribution of each of the 26 population groups to the ancestry of the donor from whom the cells were obtained.

Results: Using diverse single-cell datasets with matched whole-genome sequencing data, we show that scAI-SNP is robust to the sparsity of single-cell data, can accurately and consistently infer ancestry from samples derived from diverse types of tissues and cancer cells, and can be applied to different modalities of single-cell profiling assays, such as single-cell RNA-seq and single-cell ATAC-seq.

Discussion: Finally, we argue that ensuring that single-cell atlases represent diverse ancestry, ideally alongside race and ethnicity, is ultimately important for improved and equitable health outcomes by accounting for human diversity.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00029-4.

Background: Blood proteome analysis is becoming increasingly popular in veterinary research because many animal models have been used to study a range of human diseases. Most of the commercial high-abundance protein (HAP) depletion kits are optimized for human serum, whereas proteins in animal serum may not be present in human serum or may be present at different abundances. There are no previous studies investigating the efficacy of using various HAP kits for proteome analysis of animal serums.

Method: We used three commercial serum abundant protein depletion (SAPD) kits (i.e., ion exchange-based Norgen kit (ProteoSpin™), antibody-based Thermo Albumin Depletion Kit (Pierce™), solubility-based kit (Minute™), and a cost-effective method (i.e., perchloric acid (PerCA) HAP precipitation) to assess their effectiveness to process serums from five different species (i.e., mouse, chicken, dog, goat, and guinea pig). Protocols of the commercial kits were adopted from manufacturers' guidelines with minor modifications for optimized performance. Following HAP depletion, proteins from all species were digested using a Trypsin/Lys-C enzyme mix, desalted, and subjected to label-free quantitative bottom-up proteomics analysis via liquid chromatography-tandem mass spectrometry (LC-MS/MS). The raw data were processed using the Andromeda search engine integrated into MaxQuant, and peptide identification was performed by searching against the UniProt-reviewed protein database. Advanced bioinformatics tools were employed to facilitate data analysis and visualization, ensuring comprehensive interpretation of the depletion efficiency and comparative performance of the methods across species.

Result: We determined their capabilities of protein identification (Norgen kit > Minute kit > PerCA precipitation > Thermo kit), depletion efficiencies of HAPs (Minute kit > Norgen kit > PerCA precipitation > Thermo kit), and cost-effectiveness (PerCA precipitation > Minute kit > Norgen kit > Thermo kit). Our results show that the PerCA precipitation method, which is > 20 times cheaper than commercial kits, outperforms other methods in depleting HAPs, especially in mouse serum. While Norgen kit excels in mouse and goat serum, the PerCA precipitation method offers broader applicability and reveals unique low abundant proteins. Protein pathway analysis highlights distinct biological processes affected by different depletion methods.

Discussion: Overall, our studies provide valuable insights into protein depletion techniques, with the PerCA depletion method emerging as a cost-effective and versatile option for proteomics research across various serums.

Background: The isolation of neuron-derived extracellular vesicles (nEVs) from biofluids offers the potential to discover novel biomarkers to aid in diagnosis and treatment of psychiatric and neurodegenerative diseases. A few studies have used anti-NCAM antibody-bead-based immunocapture to enrich nEVs from plasma, some with little method validation. We therefore examined in detail this method for nEV enrichment.

Methods: EVs were isolated from SH-SY5Y cell-conditioned media by precipitation, or from plasma using size exclusion chromatography. EVs were characterised using nanoparticle tracking analysis (NTA), transmission electron microscopy (TEM) and immunoblot analysis. SH-SY5Y-EVs were incubated with anti-NCAM immunocapture beads and examined by flow cytometry, immunoblot analysis and scanning electron microscopy (SEM). Immunocaptured plasma-derived EVs were examined using a sensitive NCAM ELISA, SEM and qPCR for miRNAs.

Results: Characterisation of SH-SY5Y-derived and plasma-derived EVs revealed the expected size distributions of EVs using NTA, the presence of EV markers using immunoblot analysis, and a cup-shaped morphology using TEM. Anti-NCAM beads, but not anti-L1CAM or IgG beads, captured NCAM-positive SH-SY5Y-EVs as shown by flow cytometry and immunoblot analysis. Both SH-SY5Y and plasma-derived EVs were visualised on the surface of anti-NCAM immunocapture beads using SEM. A sensitive NCAM ELISA detected NCAM antigen in plasma-derived EVs immunocaptured on anti-NCAM beads. qPCR analysis of plasma-derived EVs detected many miRNAs in total plasma-EVs with high expression of hsa-miR-16-5p, hsa-miR-451a and hsa-miR-126-3p. However, only between two and seven miRNAs were detected in EVs captured on anti-NCAM-beads from three blood donors. Finally, tissue distribution analysis of miRNAs from plasma-derived EVs on anti-NCAM beads revealed that these miRNAs are enriched in tissues or organs such as blood vessels, lung, bone, thyroid and heart, but were not enriched for brain-derived miRNAs.

Discussion: This study indicates that anti-NCAM beads can efficiently enrich NCAM-positive EVs from cell culture conditioned media. However, nEV levels in small volumes of plasma are possibly too low to enable efficient anti-NCAM immunocapture for subsequent miRNA analysis. Other neuron-specific markers with high expression levels on nEVs are therefore required for processing patient samples where plasma volumes are low, and to allow efficient isolation of nEVs in clinical studies for subsequent cargo analysis.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00034-7.

Background: Cells employ myriad regulatory mechanisms to maintain protein homeostasis, termed proteostasis, to ensure correct cellular function. Dysregulation of proteostasis, which is often induced by physiological stress and ageing, often results in protein aggregation in cells. These aggregated structures can perturb normal physiological function, compromising cell integrity and viability, a prime example being early onset of several neurodegenerative diseases. Understanding aggregate dynamics in vivo is therefore of strong interest for biomedicine and pharmacology. However, factors involved in formation, distribution and clearance of intracellular aggregates are not fully understood.

Methods: Here, we report an improved methodology for production of fluorescent aggregates in model budding yeast which can be detected, tracked and quantified using fluorescence microscopy in live cells. This new openly-available technology, iPAR (inducible Protein Aggregation Reporter), involves monomeric fluorescent protein reporters fused to a ∆ssCPY* aggregation biomarker, with expression controlled under the copper-regulated CUP1 promoter.

Results: Monomeric tags overcome challenges associated with non-physiological reporter aggregation, whilst CUP1 provides more precise control of protein production. We show that iPAR and the associated bioimaging methodology enables quantitative study of cytoplasmic aggregate kinetics and inheritance features in vivo. We demonstrate that iPAR can be used with traditional epifluorescence and confocal microscopy as well as single-molecule precise Slimfield millisecond microscopy. Our results indicate that cytoplasmic aggregates are mobile and contain a broad range of number of iPAR molecules, from tens to several hundred per aggregate, whose mean value increases with extracellular hyperosmotic stress.

Discussion: Time lapse imaging shows that although larger iPAR aggregates associate with nuclear and vacuolar compartments, we show directly, for the first time, that these proteotoxic accumulations are not inherited by daughter cells, unlike nuclei and vacuoles. If suitably adapted, iPAR offers new potential for studying diseases relating to protein oligomerization processes in other model cellular systems.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00023-w.

Multi-omic dataset can better characterize complex cellular signaling pathways from multiple views compared to individual omic data. However, integrative multi-omic data analysis to rank key disease biomarkers and infer core signaling pathways remains an open problem. In this study, we developed a novel graph AI model, mosGraphFlow, for analyzing multi-omic signaling graphs (mosGraphs), 2) analyzed multi-omic mosGraph datasets of Alzheimers' Disease (AD), and 3) developed a visualization tool to facilitate the visualization of identified disease associated signaling biomarkers and network. The comparison results show that the proposed model not only achieves the best classification accuracy but also identifies important AD disease biomarkers and signaling interactions. In the visualization, the signaling sources are highlighted at specific omic levels to facilitate the understanding of disease pathogenesis. The proposed model can also be applied and expanded for other multi-omic data-driven studies. The code of the model is publicly accessible via GitHub: https://github.com/FuhaiLiAiLab/mosGraphFlow.

Supplementary information: The online version contains supplementary material available at 10.1186/s44330-025-00041-8.